Innovative Methods & Data Science Program
Developing novel computational, statistical, and other data science methods to rigorously quantify associations between novel, multimodal big data and health outcomes
Evidence-based knowledge comes from good, rich data combined with rigorous analysis. Unlocking the potential of data requires both the judicious application of existing analytic tools as well as a commitment to developing new analytic techniques that address the complexities of modern health systems data. The Innovative Methods & Data Science (IMDS) program will provide a key link in the learning health systems pipeline from data acquisition to clinical insights including development of innovative computational and statistical methods and tools to ensure that they remain efficacious and equitable, as well as quality improvement during the deployment process.
IMDS launched in November 2022 as a program within CLHSS in partnership with the School of Public Health Division of Biostatistics and is in the process of establishing operations and services. The initial goals of IMDS are to:
- Support the development of innovative computational methods for accountable, interpretable, impactful and fair analysis of health system data;
- Partner with CLHSS researchers to augment AI/ML tools with statistical and computational insights;
- Provide access to expert consultation on modern study designs and data analytic methods relevant to health systems data for CLHSS-affiliated faculty, staff, and students.
If you’re interested in engaging with IMDS, please email [email protected].
IMDS People
IMDS Leadership
Scientific co-Director, Julian Wolfson, PhD
Scientific co-Director, Rui Zhang, PhD
IMDS Core Member
Project Manager, Matt Loth, PhD
IMDS Full Members
Jue Hou, PhD
Jared Huling, PhD
Erich Kummerfeld, PhD
Lianne Siegel, PhD
Ju Sun, PhD
Jinhua Wang, PhD
IMDS Recent News
Developing a Framework and Metrics to Measure Fairness of Risk Prediction Models. PhD student Solvejg Wastvedt developed a framework and new metrics to measure fairness of risk prediction models with support from IMDS. Her work addresses issues with existing methods that make them difficult to apply to healthcare.